ADR-152: WiFi-Pose SOTA 2026 intake — WiFlow-STD benchmark, Rust integrations, ADR-153 802.11bf layer, efficiency frontier (#1008)

* feat(calibration): NodeGeometry transceiver-geometry recording (ADR-152 §2.1.1) PerceptAlign-motivated geometry capture at enrollment: per-node optional records (position, antenna orientation, inter-node distances, acquisition method) — recorded when known, never required. Event-sourced via EnrollmentEvent::GeometryRecorded (latest recording wins); persisted on SpecialistBank with serde defaults so pre-ADR-152 bank JSON loads cleanly (fixture-proven, and geometry-free banks serialize byte-shape-identical to the old schema); threaded through MultiNodeMixture as data only — the learned geometry embeddings and algorithmic fusion use are §2.1.2, deliberately deferred until the ADR-151 P6 LoRA heads exist. Geometry recorded from now on means banks captured today remain usable for layout-conditioned training later — you can't retroactively add geometry to data you didn't record. 8 new tests (3 geometry, 2 anchor, 2 bank, 1 multistatic) + full-loop extension (2-node geometry, one tape-measured + one unknown, surviving the bank JSON round-trip the runtime loads from). 50/50 calibration (both feature configs) + 23 CLI tests green. Co-Authored-By: RuFlo <ruv@ruv.net> * feat(training): two-checkerboard camera↔room calibration for ADR-079 labels (ADR-152 §2.1.3) Defends the camera-supervised pipeline against PerceptAlign's "coordinate overfitting": MediaPipe keypoints were emitted in raw camera coordinates with no shared frame and no transceiver-geometry metadata — the exact label shape that memorizes deployment layout and collapses cross-layout. - scripts/calibrate-camera-room.py + calibration_lib.py: OpenCV two-checkerboard calibration → versioned bundle JSON (intrinsics, camera→room extrinsics, checkerboard spec, transceiver geometry, sha256 calibration_id). Intrinsics resolve from file > cache > multi-view computation > loud-warning 2-view fallback. - collect-ground-truth.py --calibration <bundle>: every sample gains keypoints_room (unit bearing rays from the camera center in the room frame — documented projective alignment; raw image coords preserved so training chooses), camera_origin_room, calibration_id, and the transceiver geometry stamp. Without the flag, output is byte-identical to before (tested) + a one-line ADR-152 warning. Design finding (recorded for ADR-152): a single planar checkerboard's corner grid is centrosymmetric — the reversed corner ordering fits a ghost camera pose with IDENTICAL reprojection error, so per-board flip disambiguation is mathematically ill-posed. solve_two_board_extrinsics solves the joint wall+floor set over all 4 flip combinations, where the minimum is unique — an independent reason the TWO-checkerboard method is required, beyond what PerceptAlign states. 15 headless pytest tests green (synthetic corners: extrinsics recovery incl. ghost resolution, bundle round-trip + hash stability, ray transforms w/ distortion + cross-resolution, no-calibration byte identity). Co-Authored-By: RuFlo <ruv@ruv.net> * feat(benchmarks): WiFlow-STD reproduction harness + measurement (a) results (ADR-152 §2.2) Shipped checkpoint REFUTED (0.08% PCK@20, wrong keypoint normalization); 6 reproducibility defects documented (broken imports, corrupted dataset tail with float32-max garbage that NaN-poisons fp16 BatchNorm, unreachable test phase). After repairs, retraining with upstream defaults reproduces 96.09% PCK@20 full-test / 96.61% corruption-free (published 97.25%) on RTX 5080. Claims graded MEASURED-EQUIVALENT; 2.23M params + ~0.055 GFLOPs verified. Third-party code/weights/data stay out of tree (gitignored). Co-Authored-By: claude-flow <ruv@ruv.net> * feat: ADR-152 Rust integrations + ADR-153 802.11bf protocol model - calibration: GeometryEmbedding — 32-slot permutation-invariant NodeGeometry featurization for future LoRA-head conditioning (ADR-152 §2.1.2); derived SpecialistBank::geometry_embedding() accessor; 59 tests - train: MaePretrainConfig + patchify/random-mask with UNSW measured recipe (80% masking, (30,3) patches; ADR-152 §2.3, arXiv 2511.18792); strict no-truncate/no-NaN policy; proptest properties - train: WiFlowStdModel — tch-gated port of the verified ~96%-PCK@20 WiFlow-STD architecture (ADR-152 §2.2 beyond-SOTA); ungated param formula pinned to 2,225,042; 15/17-keypoint support; 239 crate tests - hardware: ieee80211bf forward-compatibility protocol model (ADR-153): SpecProfile gates, SensingCapabilities negotiation, required ConsentMode, session FSM, SensingTransport + SimTransport + OpportunisticCsiBridge; full acceptance checklist covered; 156+4 tests - deps: ruvector bumps per ADR-152 §2.6 survey (mincut/solver 2.0.6, attention 2.1.0, gnn 2.2.0); vendor/ruvector synced to a083bd77f - docs: ADR-153 accepted; ADR-152 §2.2 status, §2.4 amendment, §2.6 added Workspace: 162 test suites green (--no-default-features); Python proof PASS. Known pre-existing flake: homecore-api env_empty_falls_back_to_defaults (unserialized env-var mutation) — untouched, follow-up. Co-Authored-By: claude-flow <ruv@ruv.net> * docs: CHANGELOG + CLAUDE.md entries for ADR-152 integrations and ADR-153 Co-Authored-By: claude-flow <ruv@ruv.net> * fix(train): repair tch-backend bit-rot — gated path compiles and tests run again Mechanical API refresh against current tch: Vec::from(Tensor) -> try_from (+ explicit flatten), numel() usize cast, Rem/div ops -> remainder() / divide_scalar_mode(floor) — the latter fixed a silent true-division bug in heatmap argmax decoding; clamp(1.0, f64::MAX) -> clamp_min (torch 2.x scalar overflow panic); petgraph EdgeRef import; missing EvalMetrics and verify_checkpoint_dir APIs that tests documented. wiflow_std roundtrip test uses safetensors (.pt _save_parameters roundtrip broken in torch 2.11 Windows). Gated: 349 passed (incl. all 20 wiflow_std); ungated: unchanged. Known pre-existing: gaussian-heatmap convention mismatch (2 tests), proof seed race under parallel threads — documented, deliberate follow-ups. Co-Authored-By: claude-flow <ruv@ruv.net> * feat(train): WiFlow-STD PyTorch->tch weight import + numerical parity proof export_to_safetensors.py maps the retrained checkpoint (295 tensors -> 248 mapped, param sum exactly 2,225,042; num_batches_tracked dropped) into a tch-loadable safetensors plus a deterministic parity fixture. Gated #[ignore] integration test loads it strictly and asserts forward-pass agreement: max abs diff 1.192e-7 on the seed-42 fixture. dump_variable_names test makes the tch name layout authoritative. Zero architecture discrepancies found. Co-Authored-By: claude-flow <ruv@ruv.net> * fix: workflow-review findings — BN gamma init, ThresholdParams serde, init docs Concurrent validation workflow (2 review lanes + adversarial verification, 13 agents): 5 confirmed findings, 3 refuted. Fixes: - wiflow_std: pin BatchNorm gamma to 1.0 (tch default draws Uniform(0,1) — silently halves activations in from-scratch training; loaded checkpoints unaffected, parity re-verified after the change) - wiflow_std: document the conv-init divergences vs the reference's effective kaiming_normal(fan_out) re-init (from-scratch dynamics only) - ieee80211bf: ThresholdParams deserialization validates via try_from so the <=100 invariant holds for untrusted payloads (+ rejection test) Benchmarks (release, ruvzen): GeometryEmbedding 1.84us/call (542k/s), MAE tokenization 7.38us/window (135k/s), 802.11bf FSM 8.9M events/s — nothing suspicious. Co-Authored-By: claude-flow <ruv@ruv.net> * docs(adr): ADR-152 §2.1.4 gate resolved — PerceptAlign repo MIT, dataset on HF Co-Authored-By: claude-flow <ruv@ruv.net> * feat(benchmarks): edge optimization measured + measurement (b) blocked + 92.9% retraction Edge optimization (ADR-152 optimize track): ONNX Runtime fp32 is the CPU latency win (3.2 ms/window, ~3.4x faster than torch, parity 2.4e-7); ORT dynamic int8 reaches 2.44 MB (paper's ~2.2 MB claim plausible only via conv-capable toolchains; -0.16pt PCK@20, +18% MPJPE, 2x slower); torch dynamic quant converts 0% of this conv-only model; fp16 halves storage free but is slower on CPU. Measurement (b) BLOCKED-ON-DATA: only 1,077 paired ESP32 windows exist (stop rule <2k). Forensic recheck of the surviving April holdout RETRACTS the ADR-079 '92.9% PCK@20' figure: constant-output model, absolute (not torso) threshold, 69 near-static frames — mean predictor scores 100% under that protocol; torso-PCK@20 is 19.1%. Corroborates PR #535. Stale citations removed from user-guide, readme-details, ADR-152 §2.1.3; no-citation rule extended to ADR-079 accuracy claims. Unblock: >=2k-window multi-pose paired session + torso-PCK re-baseline. Co-Authored-By: claude-flow <ruv@ruv.net> * docs(user-guide): corrected camera-supervised collection tutorial Step 0 CSI-rate check + session-length math (window yield = frames/20 — the May session's 8x under-delivery was a ~12 Hz CSI rate, not an aligner bug); two-checkerboard calibration step (ADR-152 §2.1.3); pose-variety and confidence guidance; torso-normalized PCK + temporal-split + pred-variance eval protocol (lessons from the 92.9% retraction); scale presets re-keyed to realistic window counts. Co-Authored-By: claude-flow <ruv@ruv.net> * feat(benchmarks): static PTQ int8 (calibrated) results + overnight capture script Conv-only static QDQ beats dynamic int8 on accuracy (PCK@20 96.61-96.63% vs 96.52%, MPJPE +10% vs +18% over fp32) at ~equal size/latency; all-ops QDQ strictly worse (int8 activations through attention glue). Entropy calibration verified bit-identical to MinMax on this data. Deployment: ONNX fp32 for speed (3.2ms), static conv-only QDQ for smallest (2.53MB). Also: scripts/overnight-empty-capture.py — segmented UDP CSI recorder for empty-room baselines (no glob collisions, detach-safe). Co-Authored-By: claude-flow <ruv@ruv.net> * feat(benchmarks): measurement (b) MEASURED — optimization transfer only, mean-pose baseline wins WiFlow-STD fine-tuned on 2,046 fresh single-room ESP32 paired windows (temporal 70/15/15, 70->540 adapter, K=17): pretrained-init 65% PCK@20 vs scratch 0% (optimization transfer) but frozen-trunk ~0% (no feature transfer), and NOTHING beats the mean-pose baseline (95.9% PCK@20 — single subject, near-static normalized coords). Honesty gates held: pred std 0.0113 (non-constant model) but mean-baseline dominance means no citable CSI->pose capability from this data. ADR-152 open question 1 answered partially; definitive answer needs multi-subject/position data. Two new aligner findings: heterogeneous csi_shape with silent zero-padding (~20%), and extractCsiMatrix's transposed shape label (frame-major data, [nSc, nFrames] label) — fixes pending. Co-Authored-By: claude-flow <ruv@ruv.net> * feat(benchmarks): efficiency sweep MEASURED — half model dominates full reference Compact WiFlow-STD variants on the same data/split/protocol: half (843,834 params, 0.38x) strictly dominates the 2.23M reference (PCK@20 96.62 vs 96.61, PCK@50 99.47 vs 99.11, MPJPE 0.00898 vs 0.0094) — the published architecture is over-parameterized for its own benchmark. quarter (338k) 96.05%; tiny (56,290 params, 1/39.5) holds 94.11% — a ~220KB fp32 edge candidate. In-domain caveats recorded; cross-domain untested. Co-Authored-By: claude-flow <ruv@ruv.net> * feat(train): compact WiFlow-STD presets in Rust + tiny edge artifact (ADR-152) WiFlowStdConfig gains half()/quarter()/tiny() mirroring the overnight sweep exactly: TcnGroupsMode (Fixed/Gcd/Depthwise), input_pw_groups, derived stride schedule and decoder-mid (all default to upstream behavior; legacy serde JSON unaffected). Param formulas pin to trained ground truth first try: 843,834 / 338,600 / 56,290; default 2,225,042 pin and 1.192e-7 parity unchanged. 248 tests green. Tiny edge artifact (tiny_edge_bench.py): ONNX fp32 = 295 KB, 0.66 ms/win (~1,500/s CPU), 94.11% PCK@20 (matches sweep clean-test exactly; parity 1.49e-7). Static int8 is a bad trade at this scale (-1.43pt, +19% MPJPE, -16% size, slower) — recorded as negative result. Export note: width-16 breaks AdaptiveAvgPool((15,1)) TorchScript export; replaced by exact mean+matmul equivalent, proven by parity. Co-Authored-By: claude-flow <ruv@ruv.net> * fix: resolve all 10 confirmed code-review findings (7-angle review, 20/20 verified) wiflow_std: min_feature_width (default 15) replaces the keypoints->stride coupling — for_keypoints(17) now provably builds the trained [2,2,2,2] graph and pools 15->17, matching the validated Python protocol (pinned by tests); param_count() total on invalid configs; random_mask returns Result and rejects non-finite/out-of-range ratios; trainer checkpoints switched to safetensors (.pt VarStore roundtrip broken on Windows torch 2.11). ieee80211bf: SBP proxy now re-triggers instances and relays reports via Action::RelaySbpReport -> SensingFrame::SbpReport (clients consume via their existing path); missed_instances reset on success = consecutive semantics; SessionTable gains a guarded SBP entry point + unknown-id drop counter; initiator-role sessions reject inbound setup/SBP requests (RejectedNotSupported) closing the idle hijack; StartSetup/StartSbp outside Idle return InvalidStateForCommand; SBP validation unified through evaluate_setup with a 1:1 SetupStatus->SbpStatus mapping. events.rs split out to honor the 500-line cap. calibration/cli: enrollment geometry now actually reaches trained banks — both production call sites attach .with_geometry; --geometry flag on train-room and POST /enroll/geometry + train-body geometry on calibrate-serve give production a recording surface; geometry-free banks log the ADR-152 §2.1.2 note. benchmarks: corruption masks committed as ground truth (unregenerable after in-place cleaning; verified bit-identical regeneration from the pristine copy) + generate_corruption_masks.py producer; _bench_common.py dedups the 5x-copied shim/evaluate/seed/remap (post-refactor PCK@20 re-verified equal to the last digit); remote scripts get the mmap patch; tiny_edge --calib validated multiple-of-64; onnx_bench --help no longer executes (and overwrote) the export — artifact restored byte-exact. Workspace: 2,963 tests passed, 0 failed; Python proof PASS. Co-Authored-By: claude-flow <ruv@ruv.net> * ci: build workspace tests without debuginfo — runner disk exhaustion The combined 38-crate debug target exceeds the GitHub runner's disk ('final link failed: No space left on device'); the same tree measured 151GB locally with full debuginfo. CARGO_PROFILE_{DEV,TEST}_DEBUG=0 shrinks the target ~5-10x; debuginfo serves no purpose in CI test runs. Co-Authored-By: claude-flow <ruv@ruv.net>
2026-08-08 20:11:43 +00:00 · 2026-06-11 17:02:23 -04:00
parent 29de574e63
commit 17471e93ff
79 changed files with 14132 additions and 117 deletions
@@ -11,7 +11,7 @@ use std::time::{Duration, Instant, SystemTime, UNIX_EPOCH};
 use tokio::net::UdpSocket;
 use wifi_densepose_calibration::{
    Anchor, AnchorLabel, AnchorQualityGate, AnchorRecorder, EnrollmentEvent, EnrollmentSession,
-    MixtureOfSpecialists, MultiNodeMixture, SpecialistBank,
+    MixtureOfSpecialists, MultiNodeMixture, NodeGeometry, SpecialistBank,
 };
 use wifi_densepose_calibration::extract::{AnchorFeature, Features};
 use wifi_densepose_core::types::CsiFrame;
@@ -226,20 +226,50 @@ pub struct TrainRoomArgs {
    /// Output specialist-bank file.
    #[arg(long, default_value = "./room-bank.json")]
    pub output: String,
+    /// Optional transceiver-geometry file: a JSON array of `NodeGeometry`
+    /// records (ADR-152 §2.1.1). Recorded into the enrollment session before
+    /// training so the bank carries the layout it was trained under.
+    #[arg(long)]
+    pub geometry: Option<String>,
 }

 /// Execute `train-room`.
+///
+/// If the enrollment session carries a transceiver-geometry snapshot (recorded
+/// at enroll time or supplied here via `--geometry`), it is threaded into the
+/// bank (ADR-152 §2.1.1); a geometry-free enrollment still trains a valid bank.
 pub async fn train_room(args: TrainRoomArgs) -> Result<()> {
    let raw = std::fs::read_to_string(&args.enrollment)
        .map_err(|e| anyhow::anyhow!("cannot read {}: {e} — run `enroll` first", args.enrollment))?;
-    let data: EnrollmentData =
+    let mut data: EnrollmentData =
        serde_json::from_str(&raw).map_err(|e| anyhow::anyhow!("invalid enrollment: {e}"))?;
    if data.anchors.is_empty() {
        bail!("no accepted anchors in {} — re-run enroll", args.enrollment);
    }

-    let bank = SpecialistBank::train(&data.room_id, &data.baseline_id, &data.anchors, now_unix())
+    if let Some(path) = &args.geometry {
+        let graw = std::fs::read_to_string(path)
+            .map_err(|e| anyhow::anyhow!("cannot read geometry {path}: {e}"))?;
+        let geometry: Vec<NodeGeometry> = serde_json::from_str(&graw).map_err(|e| {
+            anyhow::anyhow!("invalid geometry {path}: {e} (expected a JSON array of NodeGeometry records)")
+        })?;
+        data.session.record_geometry(geometry, now_unix());
+    }
+
+    let mut bank = SpecialistBank::train(&data.room_id, &data.baseline_id, &data.anchors, now_unix())
        .map_err(|e| anyhow::anyhow!("training failed: {e}"))?;
+    match data.session.geometry() {
+        Some(g) if !g.is_empty() => {
+            bank = bank.with_geometry(g.to_vec());
+            eprintln!(
+                "[train-room] geometry: {} node(s) snapshotted into the bank (ADR-152 §2.1.1)",
+                bank.geometry.len()
+            );
+        }
+        _ => eprintln!(
+            "[train-room] no transceiver geometry recorded — bank will not support geometry conditioning (ADR-152 §2.1.2)"
+        ),
+    }
    std::fs::write(&args.output, bank.to_json().map_err(|e| anyhow::anyhow!("{e}"))?)
        .map_err(|e| anyhow::anyhow!("cannot write {}: {e}", args.output))?;

@@ -456,3 +486,141 @@ async fn room_watch_multi(args: RoomWatchArgs) -> Result<()> {
    }
    Ok(())
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    fn feature(label: AnchorLabel, variance: f32, motion: f32) -> AnchorFeature {
+        AnchorFeature {
+            room_id: "t".into(),
+            label,
+            features: Features {
+                mean: 1.0,
+                variance,
+                motion,
+                breathing_score: 0.0,
+                breathing_hz: 0.0,
+                heart_score: 0.0,
+                heart_hz: 0.0,
+            },
+        }
+    }
+
+    /// Write a minimal valid enrollment file (two anchors, no geometry event).
+    fn write_enrollment(dir: &std::path::Path) -> String {
+        let data = EnrollmentData {
+            room_id: "t".into(),
+            baseline_id: "base-1".into(),
+            fs_hz: 15.0,
+            anchors: vec![
+                feature(AnchorLabel::Empty, 1.0, 0.1),
+                feature(AnchorLabel::StandStill, 10.0, 0.2),
+            ],
+            session: EnrollmentSession::new("t", "base-1", 1000),
+        };
+        let path = dir.join("enrollment.json");
+        std::fs::write(&path, serde_json::to_string(&data).unwrap()).unwrap();
+        path.to_string_lossy().into_owned()
+    }
+
+    fn trained_bank(out: &std::path::Path) -> SpecialistBank {
+        SpecialistBank::from_json(&std::fs::read_to_string(out).unwrap()).unwrap()
+    }
+
+    /// ADR-152 §2.1.1: `--geometry` records into the session and the bank
+    /// snapshots it — enrollment geometry reaches the trained bank.
+    #[tokio::test]
+    async fn train_room_threads_geometry_when_provided() {
+        let dir = tempfile::tempdir().unwrap();
+        let enrollment = write_enrollment(dir.path());
+        let geometry = vec![
+            NodeGeometry::new(1, "tape-measure").with_position(0.0, 0.0, 1.0),
+            NodeGeometry::unknown(2),
+        ];
+        let gpath = dir.path().join("geometry.json");
+        std::fs::write(&gpath, serde_json::to_string(&geometry).unwrap()).unwrap();
+        let out = dir.path().join("bank.json");
+
+        train_room(TrainRoomArgs {
+            enrollment,
+            output: out.to_string_lossy().into_owned(),
+            geometry: Some(gpath.to_string_lossy().into_owned()),
+        })
+        .await
+        .unwrap();
+
+        assert_eq!(trained_bank(&out).geometry, geometry);
+    }
+
+    /// A geometry-free enrollment still trains a valid bank (optional by
+    /// design) — it just carries no snapshot.
+    #[tokio::test]
+    async fn train_room_without_geometry_yields_geometry_free_bank() {
+        let dir = tempfile::tempdir().unwrap();
+        let enrollment = write_enrollment(dir.path());
+        let out = dir.path().join("bank.json");
+
+        train_room(TrainRoomArgs {
+            enrollment,
+            output: out.to_string_lossy().into_owned(),
+            geometry: None,
+        })
+        .await
+        .unwrap();
+
+        let bank = trained_bank(&out);
+        assert!(bank.geometry.is_empty());
+        assert!(bank.presence.is_some(), "bank still trains without geometry");
+    }
+
+    /// Geometry recorded at enroll time (in the session event log) is picked up
+    /// without the `--geometry` flag.
+    #[tokio::test]
+    async fn train_room_uses_session_geometry() {
+        let dir = tempfile::tempdir().unwrap();
+        let geometry = vec![NodeGeometry::new(3, "floor-plan").with_position(1.0, 2.0, 1.5)];
+        let mut session = EnrollmentSession::new("t", "base-1", 1000);
+        session.record_geometry(geometry.clone(), 1000);
+        let data = EnrollmentData {
+            room_id: "t".into(),
+            baseline_id: "base-1".into(),
+            fs_hz: 15.0,
+            anchors: vec![
+                feature(AnchorLabel::Empty, 1.0, 0.1),
+                feature(AnchorLabel::StandStill, 10.0, 0.2),
+            ],
+            session,
+        };
+        let epath = dir.path().join("enrollment.json");
+        std::fs::write(&epath, serde_json::to_string(&data).unwrap()).unwrap();
+        let out = dir.path().join("bank.json");
+
+        train_room(TrainRoomArgs {
+            enrollment: epath.to_string_lossy().into_owned(),
+            output: out.to_string_lossy().into_owned(),
+            geometry: None,
+        })
+        .await
+        .unwrap();
+
+        assert_eq!(trained_bank(&out).geometry, geometry);
+    }
+
+    #[tokio::test]
+    async fn train_room_rejects_invalid_geometry_file() {
+        let dir = tempfile::tempdir().unwrap();
+        let enrollment = write_enrollment(dir.path());
+        let gpath = dir.path().join("geometry.json");
+        std::fs::write(&gpath, r#"{"not":"an array"}"#).unwrap();
+
+        let err = train_room(TrainRoomArgs {
+            enrollment,
+            output: dir.path().join("bank.json").to_string_lossy().into_owned(),
+            geometry: Some(gpath.to_string_lossy().into_owned()),
+        })
+        .await
+        .unwrap_err();
+        assert!(err.to_string().contains("invalid geometry"), "{err}");
+    }
+}